Zr Stabilized Ti5 Te4 -Type Hafnium Telluride
Meng He, Arndt Simon, and Viola Duppel
Max-Planck-Institut für Festkörperforschung, Heisenbergstraße 1, D-70569 Stuttgart, Germany
Reprint requests to Prof. Dr. Arndt Simon. Fax: +49 711 689 1642. E-mail: [email protected]
Z. Naturforsch. 60b, 284 – 288 (2005); received December 14, 2004
A new Ti5 Te4 -type compound, Zr stabilized hafnium telluride was discovered and characterized
with electron crystallography. Microanalyses and preparative work show that Zr stabilizes Ti5 Te4 type hafnium telluride and destabilizes the competing phase, Hf3 Te2 .
Key words: Hafnium, Zirconium, Ti5 Te4 -Type Structure, Electron Crystallography
Introduction
Hafnium and zirconium are commonly thought
to be very similar in chemistry. However, they
are surprisingly dissimilar in the chemistry of
metal-rich chalcogenides. For instance, Hf 2 S [1]
and Hf2 Se [2] are comprised of hexagonal closepacked double metal atom layers spaced by sulfur
or selenium while both Zr 2 S [3] and Zr 2 Se [4]
crystallize in a Ta2 P-type structure [5]. Zr 2 Te [6]
and Hf2 Te [7] adopt Sc 2 Te- [8] and Nb 2 Se-type
structures [9], respectively. Hf 3 Te2 [10] has a layered structure consisting of bcc metal slabs sandwiched by Te atoms, but no counterpart is reported
for zirconium tellurides. Ti 5 Te4 -type [11] Zr 5 Te4 was
identified a long time ago [12] and recently structurally
characterized [13, 14]. Hf 5 Te4 adopting the same structure type was mentioned only [15, 16] by citing unpublished work.
In the Ti5 Te4 -type structure, M6 X8 clusters (M =
early transition metal, X = metalloid or post transition
metal) [17] are trans-condensed into one-dimensional
bcc metal columns which are further connected with
each other via metal-metal bonding (see Fig. 1). More
than 10 Ti5 Te4 -type compounds have been reported up
to now. It was found that this structure type is adopted
only when the valence electron count (VEC) [18] and
atomic sizes [19] meet certain criteria. We predicted
Ti5 Te4 -type Hf5 Te4 to exist [19] because the abovementioned criteria are met in this case. However, it
turns out that Ti 5 Te4 -type hafnium telluride exists only
with the stabilization by zirconium. Here we report the
synthesis and characterization of this phase.
Experimental Section
Fig. 1. A projection of the Ti5 Te4 -type structure along the
tetragonal c axis is shown at the left side. The part marked
with a dashed circle in the projection represents a single
1 [M X ] column (M = early transition metal, X = metalloid
∞ 5 4
or post transition metal), which is shown in perspective view
at the right. Filled small circles depict early transition metal
atoms, open ones metalloid or post transition metal atoms.
Two crystallographic sites for early transition metal atoms
are denoted as M1 and M2, respectively.
Synthesis: Hf powder (99.6%, including 2 — 3.5% Zr,
325 mesh, Johnson Matthey), Zr rod (99.2+%, including
≤ 4.5% Hf, 12.7 mm in diameter, Johnson Matthey) and Te
(99.997+%, Preussag) were used as starting materials. Zr was
lathed into small pieces before being used. HfTe2 was first
prepared in evacuated silica ampoules and then further reduced with Hf and Zr in sealed tantalum tubes which were
protected by a silica tube by adding some silica wool. When
Zr pieces were used, a short tantalum rod was added and the
ampoules were repeatedly shaken during the heat treatment
to crush Zr pieces and to homogenize the sample. The ampoules were heated at 1000 ◦C for about 40 days and then
quenched in cool water.
c 2005 Verlag der Zeitschrift für Naturforschung, Tübingen · http://znaturforsch.com
0932–0776 / 05 / 0300–0284 $ 06.00 M. He et al. · Zr Stabilized Ti5 Te4 -Type Hafnium Telluride
285
TEM and EDXS: TEM observations and EDXS measurements were performed with a Philips CM30/ST electron microscope which was operated at 300 kV and equipped with
an energy dispersive X-ray spectroscopy system (Noran, Silicon detector). High resolution images and selected area electron diffraction (SAED) patterns were recorded with a Slow
Scan CCD Camera (Gatan, 1024 × 1024 pixels). The program Crisp [20] was used to process high resolution images.
X-ray powder diffraction: X-ray measurements were
made on a Stoe StadiP system equipped with an incident
beam curved germanium monochromator. Samples were
contained in glass capillaries of 0.2 mm diameter. Data were
collected in Debye-Scherrer geometry with a linear positionsensitive detector (2θ range 4◦ ).
Results and Discussion
A sample prepared with stoichiometric quantities of
starting materials (Hf:Te = 5:4) shows strong peaks besides reflections of Hf3 Te2 on the powder diffraction
pattern. According to the equation proposed in our previous publication [19], the unit cell volume of the hypothetical Hf5 Te4 phase is predicted to be 428.3 Å3 .
Supposing the ratio of a/c is the same as that of
Zr5 Te4 , the lattice constants of the hypothetical Hf 5 Te4
should be a = 10.63 Å and c = 3.79 Å. Most of the additional peaks observed on the powder pattern can be
assigned to this unit cell. The lattice constants refined
from powder diffraction data are a = 10.674(3) Å and
c = 3.754(1) Å, in good agreement with the predicted
values.
The sample has been further examined with TEM
and EDXS. Thin Hf 3 Te2 platelets can be easily identified based on the diffraction patterns. EDXS measurements on 14 Hf 3 Te2 crystals show an average composition of 58.0% Hf, 1.7% Zr and 40.4% Te, the highest Zr content measured being 2.5%. Ti 5 Te4 -type crystals have also been identified by SAED. These crystals are chunky blocks with irregular shape. In most
cases, it is difficult to find a thin enough area for high
resolution observations, and even SAED patterns can
be of poor quality. A significant enrichment of Zr in
these crystals has been observed. Microanalyses made
on seven Ti5 Te4 -type crystals gave an average composition of 48.6% Hf, 7.9% Zr and 43.5% Te, the lowest Zr content being 4.4%. A SAED pattern along the
[001] direction and the corresponding high resolution
image are shown in Figs 2a and b, respectively. After
lattice averaging, compensating for the contrast transfer function (CTF) and imposing the crystallographic
Fig. 2. Selected area electron diffraction pattern (a) and corresponding high resolution image (b) of Ti5 Te4 -type hafnium
telluride with electron beam parallel to [001]. (c) Reconstructed image after lattice averaging, compensating for the
contrast transfer function (CTF) and imposing the crystallographic symmetry in projection P4. A projection of the structure was superimposed on (c) to highlight the features of the
reconstructed image.
symmetry in projection P4, the reconstructed image
(Fig. 2c) can be interpreted directly as a projection of
the Ti5 Te4 -type structure along [001]. The coordinates
of the metal atoms in the ab plane are measured from
this image by searching the local maxima of the po-
M. He et al. · Zr Stabilized Ti5 Te4 -Type Hafnium Telluride
286
Table 1. Fractional coordinates of Zr stabilized Ti5 Te4 -type
hafnium telluride.
Atoms
Site
x
y
z
Hf1
2a
0
0
0
Hf2
8h
0.297
0.375
0
Te
8h
0.062
0.260
0
Note: The SAED pattern and high resolution image were taken from
a crystal of composition 42.0% Hf, 13.7% Zr and 44.3% Te, as determined by EDXS.
Table 2. Interatomic distances (Å) in Zr stabilized Ti5 Te4 type hafnium telluride.
Hf 1
−2 Hf 1
−8 Hf 2
−4 Te
3.75
3.16
2.85
Hf 2
−2 Hf 1
−2 Hf 2
−2 Hf 2
−2 Hf 2
− Te
−2 Te
−2 Te
3.16
3.41
3.60
3.75
2.79
2.80
2.81
Te
−Hf 1
−Hf 2
−2 Hf 2
−2 Hf 2
−2 Te
−2 Te
−4 Te
−2 Te
2.85
2.79
2.80
2.81
3.75
4.03
4.19
4.44
tential. The potential distribution of Te is smeared out
and its coordinates cannot be determined reliably in the
same way. Inspection of other structures of the same
type reveals that nonmetal atoms are always in contact with five M2 atoms at nearly the same distances
and one M1 atom at a slightly longer distance. The coordinates of Te in the ab plane have been determined
with this geometrical restriction. The shortest distance
between Te and Hf2 in the projection is only about
2 Å, approaching the point resolution of the microscope, 1.9 Å. The coordinates along the c axis were
fixed at 0 or 1/2 from the geometrical considerations.
Attempts were not made to refine the structure on basis
of the electron diffraction data. A detailed description
of processing high resolution images with the program
Crisp can be found in a previous publication [21]. The
atomic coordinates are listed in Table 1. Supposing the
lattice constants of the crystal are the same as those
refined from X-ray powder diffraction data, the interatomic distances were calculated (Table 2).
Another phase was identified by TEM. The composition averaged over 8 EDXS measurements is 36.9%
Hf, 3.0% Zr and 60.1% Te. The metal to nonmetal ratio
in this phase suggests a compound Hf 2 Te3 . It is worth
noting that the Zr content of this phase is in between
that of Hf3 Te2 and the Ti5 Te4 -type phase. SAED patterns shown in Fig. 3 indicate that this phase is closely
related to HfTe2 . The strong reflections can be indexed
Fig. 3. Selected area electron diffraction patterns of Hf2 Te3 .
The indexing is based on a HfTe2 -related unit cell.
with a unit cell comparable with that of HfTe 2 . The
positions of other reflections show that the structure
of this phase is quite different from Hf 2 Se3 [22] and
Zr2 Te3 (Zr1.3 Te2 ) [23]. Besides sharp Bragg reflections
diffuse scattering along c ∗ is observed which, however,
is not investigated further.
The significant enrichment of Zr in Ti 5 Te4 -type
crystals raises the suspicion that it is a Zr stabilized
phase, and introducing Zr intentionally into the starting
materials may improve the yield of this phase. A sam-
M. He et al. · Zr Stabilized Ti5 Te4 -Type Hafnium Telluride
Fig. 4. Experimental powder diffraction patterns of samples
with nominal composition Hf4.5 Zr0.5 Te4 (a) and Hf5 Te4 (c)
(Cu-Kα1 , 1 min/0.1◦ in 2θ ). Simulated patterns of Ti5 Te4 type hafnium telluride (b) and Hf3 Te2 (d) are also given for
comparison. Peaks marked with “+” and “*” result from impurities of Hf2 Te3 and monoclinic HfO2 , respectively.
ple with nominal composition Hf: Zr: Te = 4.5: 0.5: 4
was therefore prepared. The X-ray powder diffraction
pattern of this sample is shown in Fig. 4 together with
that of the stoichiometric one. It can be seen that Zr
intentionally added to the starting materials improves
the yield of the Ti 5 Te4 -type phase greatly. On the other
hand, the amount of Hf 3 Te2 in the product is negligible. Clearly, Zr stabilizes the Ti 5 Te4 -type structure
[1] H. F. Franzen, J. Graham, Z. Kristallogr. 123, 133
(1966).
[2] H. F. Franzen, J. G. Smeggil, B. R. Conard, Mater. Res.
Bull. 2, 1087 (1967).
[3] X. Yao, H. F. Franzen, J. Less-Common Met. 142, 27
(1988).
[4] H. F. Franzen, L. J. Norrby, Acta Crystallogr. B24, 601
(1968).
[5] A. Nylund, Acta Chem. Scand. 20, 2393 (1966).
[6] G. Örlygsson, B. Harbrecht, Inorg. Chem. 38, 3377
(1999).
[7] B. Harbrecht, M. Conrad, T. Degen, R. Herbertz, J. Alloys Compds. 255, 178 (1997).
[8] P. A. Maggard, J. D. Corbett, Angew. Chem. Int. Ed.
Engl. 36, 1974 (1997).
[9] B. R. Conard, L. J. Norrby, H. F. Franzen, Acta Crystallogr. B25, 1729 (1969).
[10] R. L. Abdon, T. Hughbanks, Angew. Chem. Int. Ed.
Engl. 33, 2328 (1994).
[11] F. Grønvold, A. Kjekshus, F. Raaum, Acta Crystallogr.
14, 930 (1961).
287
which consists of one-dimensional bcc metal columns,
and destabilizes the Hf 3 Te2 -type structure built up
from two-dimensional bcc metal slabs.
It has been found that Hf atoms prefer sites allowing
for more metal-metal bonding in the quasi-binary system Hf-Zr-P [24]. Comparing the structures of Hf 2 Te
and Zr2 Te, Harbrecht and coworkers [6] found that
metal atoms in the former have enhanced metal-metal
bonding compared to the latter. The preference of Hf
to form more metal-metal bonding was explained by
the greater expansion of the 5d orbitals compared to
the 4d orbitals of Zr [25]. The Hf 3 Te2 -type structure allows for more metal-metal contacts than the
Ti5 Te4 -type structure does, so it is preferred by Hf, and
the Ti5 Te4 -type structure is adopted only when sufficient Zr is present to stabilize it. Obviously, only a
small amount of Zr is enough to stabilize the Ti 5 Te4 type structure or destabilize the Hf 3 Te2 -type structure: The highest Zr content in Hf 3 Te2 and the lowest Zr content in the Ti 5 Te4 -type structure observed
by EDXS are 2.5% and 4.4%, respectively. The finding that the Ti5 Te4 -type structure collects Zr from
the starting materials may be exploited to purify Hf
metal and, hence, may help to solve a long-standing
problem.
Acknowledgement
M. He thanks Max Planck Society for the financial
support.
[12] L. Brattås, A. Kjekshus, Acta Chem. Scand. 25, 2350
(1971).
[13] R. de Boer, E. H. P. Cordfunke, P. van Vlaanderen,
D. J. W. Ijdo, J. R. Plaisier, J. Solid State Chem. 139,
213 (1998).
[14] G. Örlygsson, B. Harbrecht, Z. Kristallogr. 214, 5
(1999).
[15] R. L. Abdon, T. Hughbanks, J. Am. Chem. Soc. 117,
10035 (1995).
[16] C. C. Wang, R. L. Abdon, T. Hughbanks, J. Reibenspies, J. Alloys Compds. 226, 10 (1995).
[17] A. Simon, Angew. Chem. Int. Ed. Engl. 20, 1 (1981).
[18] M. Bouayed, H. Rabaâ, A. Le Beuze, S. Députier,
R. Guérin, J. Y. Saillard, J. Solid State Chem. 154, 384
(2000).
[19] M. He, A. Simon, V. Duppel, Z. Anorg. Allg. Chem.
630, 535 (2004).
[20] S. Hovmöller, Ultramicroscopy 41, 121 (1992).
[21] T. E. Weirich, R. Ramlau, A. Simon, S. Hovmöller,
X. D. Zou, Nature (London) 382, 144 (1996).
288
[22] I. M. Schewe-Miller, V. G. Young Jr., J. Alloys Compds. 216, 113 (1994).
[23] C. C. Wang, C. Eylem, T. Hughbanks, Inorg. Chem. 37,
390 (1998).
M. He et al. · Zr Stabilized Ti5 Te4 -Type Hafnium Telluride
[24] H. Kleinke, H. F. Franzen, J. Solid State Chem. 136,
221 (1998).
[25] H. Kleinke, H. F. Franzen, Angew. Chem. Int. Ed. Engl.
35, 1934 (1996).